Progress in deciphering the genotype to phenotype relationship of the presenilin form of the familial AD provides a unique opportunity to elucidate a major unresolved issue in AD in general: how gamma-secretase generates the Abeta42 peptide which accumulates early in the course of all forms of he disease. Mutations in PS1 and PS2 selective enhance Abeta42 production in cultured cells, transgenic mice and human brain tissue, and this increase has been documented presymptomatically in plasma and primary cells from mutant gene carriers. We recently found that both PS1 and PS2 form stable complexes with APP in the ER and early Golgi and that mutant PS increases Abeta42 in these sites. Therefore, our central hypothesis of this project is that PS mutations physically alter the interaction of PS with APP in a way that allows increased proteolysis of the latter by a gamma-secretase cleaving specifically at residue 42, and that understanding how this occurs will shed light on an early and invariant feature (Abeta42 deposition) of the common "sporadic" form of AD. To address this hypothesis, we will carry out 4 Aims: 1) using stable cell lines and PS transgenic mice from Cores B and C, define the regions of APP and PS that interact, how mutations change the interaction and whether additional proteins (e.g., delta-catenin) participate in the complexes; 2) use cell biological and pharmacological methods to define all of the principal subcellular loci for Abeta40 and Abeta42 production and how PS mutations differentially effect these; 3) establish the structural requirements within APP for the gamma secretase cleavages and whether the proteases act by direct intramembranous endoproteolysis at 40 (and 42) or by a "2 hit" mechanism analogous to that recently described for SREBP; and 4) identify 42- and 40-specific gamma-secretase inhibitors and use them to characterize and ultimately purify the cognitive protease(s). All 4 Aims are based on strong preliminary data, and they will involve extensive collaborations with the other 3 PI's of this Program. The results should provide information relevant in general to the proteolytic processing of transmembrane proteins and in particular to the mechanism of Abeta42 generation, an emerging therapeutic target in Alzheimer's disease.